Image processing device and method for determining a similarity between two images

Patent metadata

Field	Value
Publication number	US-9076066-B2
Application number	US-201314100012-A
Country	US
Kind code	B2
Filing date	Dec 8, 2013
Priority date	Dec 17, 2012
Publication date	Jul 7, 2015
Grant date	Jul 7, 2015

Abstract

A computerized method determines a similarity between a first image and a second image. The first image is converted into a first grayscale image, and the second image is converted into a second grayscale image. A first feature vector of the first grayscale image and a second feature vector of the second grayscale image are extracted. A similarity value is calculated indicating the similarity between the first image and the second image according to the first feature vector and the second feature vector. If the similarity value is greater than or equal to the predetermined threshold, the first image is similar to the second image and a determination result is outputted denoting the first image is similar to the second image.

First claim

What is claimed is:

1. A computerized method for determining a similarity between a first image and a second image using an image processing device, the method comprising:
converting the first image into a first grayscale image and the second image into a second grayscale image;
extracting a first feature vector of the first grayscale image and a second feature vector of the second grayscale image, wherein the first feature vector comprises a set of first proportion values, each first proportion value of the first grayscale image is a ratio of a number of pixels having a grayscale value to a total number of pixels, the second feature vector comprises a set of second proportion values, and each second proportion value of the second grayscale image is a ratio of a number of pixels having a grayscale value to a total number of pixels, and extracting a first feature vector of the first grayscale image and a second feature vector of the second grayscale image comprises: dividing the first and second grayscale images into n grayscale histograms, and extracting a first feature vector V a ={g a0 , g a1 , g a2 , g a3 . . . g an } and a second feature vector V b ={g b0 , g b1 , g b2 , g b3 . . . g bn } wherein n=2 m −1, m is a natural number, g ai (0≦i≦n) is a ratio of a number of pixels with a grayscale value of i to the total number of the pixels of the first grayscale image, g bj (0≦j≦n) is a ratio of a number of pixels with a grayscale value of j to the total number of the pixels of the second grayscale image;
calculating a similarity value indicating the similarity between the first image and the second image according to the first feature vector and the second feature vector; and
determining whether the first image is similar to the second image by comparing the similarity value with a predetermined threshold and outputting a determination result, wherein if the similarity value is greater than or equal to the predetermined threshold, the first image is similar to the second image and the outputted determination result denotes the first image is similar to the second image.

2. The method of claim 1 , wherein the similarity value is obtained according to a formula of:

S
⁡

(

Ia
,
Ib

)

=

1

D
⁡

(

Ia
,
Ib

)

+
1

,

wherein

D
⁡

(

Ia
,
Ib

)

=

∑

i
=
0

n

⁢

⁢



g
ai

-

g
bi



.

3. The method of claim 2 , wherein the similarity value S(Ia, Ib) is greater than 0 and less than or equal to 1, and the larger the similarity value S(Ia, Ib), the more the first image is similar to the second image.

4. The method of claim 3 , wherein if the similarity value is less than the predetermined threshold, the first image is not similar to the second image and the outputted determination result denoting the first image is not similar to the second image.

5. An image processing device to determine a similarity between a first image and a second image, the image processing device comprising a storage, a processor, and one or more programs stored in the storage and executed by the processor, the one or more programs comprising:
a conversion module converting the first image into a first grayscale image and the second image into a second grayscale image;
a feature extraction module extracting a first feature vector of the first grayscale image and a second feature vector of the second grayscale image, wherein the first feature vector comprises a set of first proportion values, each first proportion value is a ratio of a number of pixels having a grayscale value to a total number of pixels of the first grayscale image, the second feature vector comprises a set of second proportion values, and each second proportion value is a ratio of a number of pixels having a grayscale value to a total number of pixels of the second grayscale image, and wherein the feature extraction module divides the first and second grayscale images into n grayscale histograms, and extracts a first feature vector V a ={g a0 , g a1 , g a2 , g a3 . . . g an } and a second feature vector V b ={g b0 , g b1 , g b2 , g b3 . . . g bn } wherein n=2 m −1, m is a natural number, g ai (0≦i≦n) is a ratio of a number of pixels with a grayscale value of i to the total number of the pixels of the first grayscale image, g bj (0≦j≦n) is a ratio of a number of pixels with a grayscale value of i to the total number of the pixels of the second grayscale image;
a similarity calculating module calculating a similarity value indicating the similarity between the first image and the second image according to the first feature vector and the second feature vector; and
a similarity determination module determining whether the first image is similar to the second image by comparing the similarity value with a predetermined threshold and outputs a determination result, wherein if the similarity value is greater than or equal to the predetermined threshold, the similarity determination module determines the first image is similar to the second image and the outputted determination result denotes that the first image is similar to the second image.

6. The image processing device of claim 5 , wherein the similarity value is calculated according to a formula of:

S
⁡

(

Ia
,
Ib

)

=

1

D
⁡

(

Ia
,
Ib

Assignees

• Fu Tai Hua Ind Shenzhen Co Ltd
• Hon Hai Prec Ind Co Ltd

Inventors

• Zhu Jian-Jian
• Xiong Yu-Kai
• Lu Xin
• Liu Hui-Feng

Classifications

• G06V10/50 Primary

  by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis · CPC title

• G06V10/758

  Involving statistics of pixels or of feature values, e.g. histogram matching · CPC title

• G06K9/6212

  Physics · mapped topic

• G06K9/4642 Primary

  Physics · mapped topic